GPU-48xCPU-480GB-2xB200 by UpCloud

GPU-48xCPU-480GB-2xB200 is a GPU (48 vCPUs, 480 GiB RAM, 2x B200) server offered by UpCloud with 48 vCPUs, 480 GiB of memory and 0 GB of storage. The last known price was 7.4345 USD per hour.
48 vCPU
480 GiB Memory
2 GPU
Spare SCore
97,241
(All-cores)
2,168
(Single-core)

Specifications #

Server Metadata #

Vendor ID
upcloud
Name
GPU-48xCPU-480GB-2xB200
Description
GPU (48 vCPUs, 480 GiB RAM, 2x B200)
Family
GPU
Series
B200
Hw Virt
Average Time To Start
84
Status
inactive
Observed At
2026-11-17T04:26:57.347792

Availability #

Region / IDSpotOndemand
Sydney #1 (AU) / au-syd1
7.4345 USD/h
9.9126 USD/h
Frankfurt #1 (DE) / de-fra1
7.4345 USD/h
9.9126 USD/h
Copenhagen #1 (DK) / dk-cph1
7.4345 USD/h
9.9126 USD/h
Madrid #1 (ES) / es-mad1
7.4345 USD/h
9.9126 USD/h
Helsinki #1 (FI) / fi-hel1
7.4345 USD/h
9.9126 USD/h
Amsterdam #1 (NL) / nl-ams1
7.4345 USD/h
9.9126 USD/h
Stavanger #1 (NO) / no-svg1
7.4345 USD/h
9.9126 USD/h
Warsaw #1 (PL) / pl-waw1
7.4345 USD/h
9.9126 USD/h
Stockholm #1 (SE) / se-sto1
7.4345 USD/h
9.9126 USD/h
Singapore #1 (SG) / sg-sin1
7.4345 USD/h
9.9126 USD/h
London #1 (GB) / uk-lon1
7.4345 USD/h
9.9126 USD/h
Chicago #1 (US) / us-chi1
7.4345 USD/h
9.9126 USD/h
New York #1 (US) / us-nyc1
7.4345 USD/h
9.9126 USD/h
San Jose #1 (US) / us-sjo1
7.4345 USD/h
9.9126 USD/h
Helsinki #2 (FI) / fi-hel2
8.1063 USD/h
11.5867 USD/h

Processor #

vCPUs
48
Hypervisor
KVM
CPU Allocation
Shared
CPU Cores
48
CPU Speed
2 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8570
CPU L1D Cache
32 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
4 MiB
CPU L2 Cache Total
192 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
16 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, arch_perfmon, rep_good, nopl, xtopology, cpuid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pdcm, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, cpuid_fault, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx_vnni, avx512_bf16, wbnoinvd, arat, avx512vbmi, umip, pku, ospke, waitpkg, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, bus_lock_detect, cldemote, movdiri, movdir64b, fsrm, md_clear, serialize, tsxldtrk, amx_bf16, avx512_fp16, amx_tile, amx_int8, arch_capabilities
Ecpus
44.8
Scalability
93.33

System Resources and Accelerators #

Memory
Memory Amount
480 GiB
Memory Amount Actual
480 GiB
Accelerators
Accelerator Count
2
Accelerator Memory Min
192 GiB
Accelerator Memory Total
384 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Blackwell
Accelerator Model
B200
Accelerators
  • NVIDIA Blackwell NVIDIA B200 (Memory amount: 183359, Firmware version: 595.71.05, BIOS version: 97.00.77.00.1E, Clock rate: 1965)
  • NVIDIA Blackwell NVIDIA B200 (Memory amount: 183359, Firmware version: 595.71.05, BIOS version: 97.00.77.00.1E, Clock rate: 1965)
Storage
Storage Size
0 GB
Storages
    Compatible Storage IDs
    maxiops, standard, hdd
    Network
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    71680 GB/month
    IPv4
    1

    CPU and System Topology #

    Server Description #

    High-density GPU acceleration with dual Blackwell GPUs and massive memory for advanced AI inference and parallel computing workloads.

    UpCloud GPU-48xCPU-480GB-2xB200 is a high-performance GPU-accelerated server built on the x86_64 architecture, featuring an Intel Xeon 8570 processor with 48 physical cores and 48 vCPUs. It is equipped with 480.0 GB of system memory and two NVIDIA Blackwell B200 GPUs delivering 384 GB of VRAM, though it includes no local storage. Benchmarks show top-tier performance in Passmark CPU tests, multi-threaded decompression, and large-block memory read bandwidth, alongside strong multi-core processing. This instance is optimized for demanding workloads such as large language model inference, deep learning, and high-throughput database operations.

    Economics #

    Average Price per Region #

    Prices per Zone #

    Lowest Prices #

    Performance #

    Workload Profiles #

    5.6Score
    Precomputed compound score for Cache Intensive workloads. A weighted average (geometric mean) of benchmark scores compared to their medians: score = ∏ (x_i / m_i)^(w_i / Σw). The score of 1.0 represents a synthetic baseline server with the median performance of each component benchmark; 0.5 means roughly half the performance; and 2.0 means twice the performance of that reference profile. Component weights: 50% Redis RPS (pipeline=1, SET), 20% Redis RPS (pipeline=16, SET), 10% PassMark Memory Mark (composite), 10% Memory bandwidth (read, 16 MB ~ L3), 10% PassMark single-thread CPU. Rationale for component selection: In-memory key-value store workload, mixing direct Redis performance metrics with memory speed and latency benchmarks, and single-core CPU performance profiles.
    Component Score Weight Impact
    Raw Reference Normalized Target Actual
    Redis RPS (pipeline=1, SET) 16,386,891 ops/sec 1,933,252 ops/sec 8.4 50.00% 50.00% +190.00%
    Redis RPS (pipeline=16, SET) 86,246,686 ops/sec 11,814,424 ops/sec 7.2 20.00% 20.00% +48.00%
    PassMark Memory Mark (composite) 2,401 2,408 1 10.00% 10.00% 0.00%
    Memory bandwidth (read, 16 MB ~ L3) 1,156,334 MB/sec 108,674 MB/sec 11 10.00% 10.00% +27.00%
    PassMark single-thread CPU 3,236 Mops/s 2,365 Mops/s 1.3 10.00% 10.00% +2.70%

    GPU Bandwidth and Latency #

    Node (all GPUs)
    Node baseline directional bandwidth
    95.4 GB/s
    Node GPU-to-host bandwidth
    111.5 GB/s
    Node host-to-GPU bandwidth
    111.3 GB/s
    Peer-to-peer
    GPU-to-GPU P2P duplex bandwidth
    770.4 GB/s
    GPU-to-GPU P2P gather bandwidth
    774.3 GB/s
    GPU-to-GPU P2P latency
    1,616 ns
    GPU-to-GPU P2P bandwidth
    774.7 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    56.1 GB/s
    Per-GPU host-to-GPU bandwidth
    55.7 GB/s
    Per-GPU host-device latency
    989 ns
    Per-GPU device memory copy bandwidth
    3193.9 GB/s
    Benchmark Environment
    Driver Version
    595.71.05
    CUDA Version (driver)
    13.2
    nvbandwidth Version
    v0.8
    GPU Count
    2
    P2P Supported
    Yes
    Efficiency
    SM/CE copy efficiency
    0.92

    Memory Bandwidth#

    Compression #

    OpenSSL #

    Geekbench Single-Core #

    Score: 2,264

    Geekbench Multi-Core #

    Score: 20,863

    Passmark CPU Scores #

    BENCHMARKSCORE
    Mark
    88,534
    Compression
    1,247,655
    Encryption
    71,927
    Extended Instructions
    132,359
    Floating Point Maths
    338,097
    Integer Maths
    288,059
    Physics
    16,556
    Prime Numbers
    905
    Single Threaded
    3,236
    String Sorting
    173,130

    Passmark Memory Scores #

    BENCHMARKSCORE
    Memory Mark
    2,401
    Database Operations
    33,289
    Memory Latency
    61
    Memory Read Cached
    29,115
    Memory Read Uncached
    8,040
    Memory Write
    7,846

    Stress-ng Stressors #

    CPU integerSingleMultiScaling
    atomic
    198.47 223.24 1.12×
    bitops
    166,166 6,872,998 41.36×
    branch
    156,700 6,466,545 41.27×
    funccall
    848.41 35,608 41.97×
    funcret
    13,949 589,517 42.26×
    hash
    199,201 8,363,828 41.99×
    intmath
    71,549 2,966,334 41.46×
    nop
    294,595 14,052,571 47.70×
    prime
    525,602 24,604,120 46.81×
    rotate
    111,470 5,274,172 47.31×
    str
    1,049,103 49,429,591 47.12×
    wcs
    771,016 35,269,326 45.74×
    CryptographySingleMultiScaling
    crypt
    3,802 151,834 39.93×
    CompressionSingleMultiScaling
    jpeg
    917.36 38,905 42.41×
    zlib
    63.37 2,696 42.55×
    HypervisorSingleMultiScaling
    clock
    45,498 188,182 4.14×
    clone
    385.71 13,389 34.71×
    context
    5,091 211,635 41.57×
    cpu-sched
    73.57 3,096 42.08×
    eventfd
    73,886 1,263,164 17.10×
    fault
    31,166 140,331 4.50×
    fork
    2,602 70,280 27.01×
    getrandom
    67,438 2,833,717 42.02×
    hrtimers
    520,126 3,063,119 5.89×
    mprotect
    595,537 15,425,469 25.90×
    mtx
    26,548,230 3,322,978,171 125.17×
    munmap
    1,919 60,525 31.53×
    null
    801,761 32,759,584 40.86×
    pipe
    727,909 23,022,719 31.63×
    priv-instr
    827,268 8,965,365 10.84×
    resched
    10,107,815 14,094,337 1.39×
    switch
    118,879 15,793,689 132.85×
    timermix
    263,279 11,953,500 45.40×
    tlb-shootdown
    9,451 175,286 18.55×
    tsc
    649,266 30,887,671 47.57×
    usersyscall
    715,085 8,614,207 12.05×
    vdso
    88,500,336 4,214,735,748 47.62×
    vfork
    22,114 159,544 7.21×
    x86syscall
    8,898,090 361,024,546 40.57×
    yield
    52,783,454 70,267,909 1.33×
    zero
    4,952,575 203,710,306 41.13×
    CPU float/SIMDSingleMultiScaling
    eigen
    17,882 749,732 41.93×
    expmath
    3,949 165,212 41.84×
    fma
    1,466,294 61,301,196 41.81×
    fp
    224.38 9,482 42.26×
    fractal
    7.2 303.69 42.18×
    hyperbolic
    4,553 192,092 42.19×
    logmath
    5,824 248,311 42.64×
    matrix
    5,344 241,973 45.28×
    matrix-3d
    788.97 19,661 24.92×
    monte-carlo
    290.33 13,868 47.77×
    mpfr
    1,967 92,824 47.20×
    powmath
    2,200 105,196 47.81×
    trig
    3,003 142,417 47.43×
    veccmp
    28,096 1,314,365 46.78×
    vecfp
    1,733 82,297 47.49×
    vecmath
    4,404 179,310 40.72×
    vecshuf
    926.16 44,125 47.64×
    vecwide
    40,190 1,909,975 47.52×
    AlgorithmsSingleMultiScaling
    bitonicsort
    10 406.99 40.70×
    bsearch
    322.93 13,321 41.25×
    bubblesort
    1.2 48 40.00×
    fibsearch
    342.66 14,478 42.25×
    heapsort
    11.2 469.99 41.96×
    hsearch
    9,443 395,340 41.87×
    insertionsort
    14.4 595.8 41.37×
    malloc
    2,993,921 55,101,032 18.40×
    mcontend
    57.1 7,432 130.16×
    mergesort
    34.4 1,626 47.26×
    qsort
    24.6 1,145 46.56×
    skiplist
    6,105 290,064 47.52×
    tree
    1.8 86.25 47.92×
    MemorySingleMultiScaling
    cache
    686,461 1,275,616 1.86×
    cachehammer
    196,011 7,309,693 37.29×
    cacheline
    33.38 3,157 94.58×
    l1cache
    362.89 15,004 41.35×
    memcpy
    497.16 20,933 42.11×
    memrate
    0.12 5.35 44.58×
    memthrash
    212.57 1,761 8.28×
    ptr-chase
    29,914,321 843,093,625 28.18×
    spinmem
    228.18 2,577 11.29×
    stream
    122.74 108,949 887.64×
    vm
    199,252 297,915 1.50×

    Stress-ng Raw Scores #

    Stress-ng Relative Multicore Performance #

    LLM Inference Speed for Prompt Processing #

    LLM Inference Speed for Text Generation #

    Static Web Server#

    Redis#

    Alternatives #

    Servers of the Same Family #

    INSTANCEvCPUsMEMORYAccelerators
    GPU-8xCPU-64GB-1xL40S 864 GiB1
    GPU-12xCPU-128GB-1xL40S 12128 GiB1
    GPU-12xCPU-240GB-1xH100 12240 GiB1
    GPU-12xCPU-128GB-2xL40S 12128 GiB2
    GPU-16xCPU-80GB-1xRTXPRO6000 1680 GiB1
    GPU-16xCPU-192GB-1xL40S 16192 GiB1
    GPU-16xCPU-192GB-2xL40S 16192 GiB2

    Similar Servers #

    INSTANCEVENDORvCPUsMEMORYAccelerators

    GPU-48xCPU-480GB-2xB200 FAQs #